ELECTRO-CHEMISTRY. 778 ELECTRO-CHEMISTRY. then connect llie t«o bars by means of a metal ■wire, a Ilow of electricity (positive electricitj' from copper to zinc, negative from zinc to cop- per) would take place between the two bars, and thus the equilibrium would be destroyed. Freed from the electrostatic attraction, the excessive zinc ions of the zinc sulphate solution would migrate toward the excessive neyative ions of the eopper sulphate solution. At the same time, more zinc, driven by the unchecked solution- tension, would go into solution, and more cop- per ions, driven by the unchecked osmotic pres- sure, would be precii)itated upon the copper bar. If we should bi'eak the connection between the bars, the former equilibrium would soon be re- established, and chemical action would cease. But as long as the connection exists, zinc goes into solution, coi)]ier is i)recipitated, and a current of ■electricity passes through tlie connecting wire; we say, "electro-chemical action is going on." Suppose now that the copper sulpliate solu- tion of the Paniell cell was replaced by a solu- tion which could take up molecules from the copper bar. but in which those molecules, instead of being transformed into ions, would form a practically undissociated compound with some other ingredient present. Tlie osmotic pressure of copper ions would thus be extremely slight and might be exceeded by the electrolytic solu- tion-tension of copper, in spite of the latter being <'omparativeIy small. It is. therefore, easy to see that under certain conditions it might be possible to reverse the electrochemical process of the cell. As a matter of fact, this is precisely what happens when the copper bar of a Daniell cell is inunersed into a concentrated solution of the cyanide of potassium. On closing the circuit, copper tlicn socs into solution, zinc is precipi- tated, and the direction of the current is reversed. This and similar phenomena were formerly de- scribed as 'anomalous,' because they seemed to contradict the assumption that zinc has a greater 'aflinity' for acids than copper. Substi- tuting the definite concept 'solution-tension' for the vague concept 'atlinity.' and adding the factor of the 'osmotic pressure of ions.' we ex- perience no difticulty in tuiderstanding efTects like the reversion of current in the nanielt cell by potassium cyanide, and many other phe- nomena. And it must be observed that the ■osmotic pressure of ions is a measurable quan- tity (see Son'riON: Dissociation), and that the solution-tension of metals, too, has now been ■calculated from directly measur.able quantities (see further below). Closely allied io the electro-chemical action of voltaic cells is the immediate precipitation of one metal by another. The precipitation of copper by iron from solutions of copper sails, and the precipitation of finely divided ('molecu- lar') silver by zinc, are familiar examples of this phenomenon, the cause of which is almost self-evident in the light of our theory. When a bar of iron, for example, is immersed in a solu- tion of copper sulphate, the solution-tension of the iron catises the appearance of some iron ions, and hence the formation of an excess of positive over negative electricity in the solution: the <'leetrostatic repulsion between the different metallic ions then tends to drive both iron ions Jind copper ions out of solution ; but the solution- tension of iron is very great, and so the iron ions remain in solution; on the other hand, the slight solution-tension of copper is readily over- come by the force of electrostatic repulsion, and so copper ions are driven out and api)ear in the state of metallic copper. An analogous exchange takes place when a metal dissolves in an aqueous solution of some acid, the solution-tension of hydrogen being overcome by the electrostatic re- pulsion between the ions of hydrogen and those of the metal. Only here the external pressure of hydrogen gas comes into play: for the solution- tension of a gas naturally depends upon its pres- sure. We have seen above that a voltaic cell may be obtained by using a dilute acid and two quantities of hydrogen gas under diflerent [iressures ; the two quantities of hydrogen act like two dill'erent metals because, owing to dilVercnce of pressure, their solution-tensions are dill'erent. Further, we see that while for certain pressures of hydro- gen the solution-tension of a given metal may be greater, for other pressures of hydrogen the solu-- tion-tension of the metal may, on the contrary, be less, than the solution-tension of hydrogen. In other words, if we should place in a vessel tilled with hydrogen vnider sufficient pressure the solution of some metallic salt, the metal might be precipitated out of the hydrogen. As a' matter of fact, this curious phenomenon has been observed in the case of several metals. If in a vessel containing hydrogen under a pressure of 18 atmospheres (270 pounds per square inch), metallic zinc is brought into contact with a solution containing per liter 12% grams of sulphuric acid and 210 grams of zinc sulphate, no action takes place. If the pressure is slightly diminished, zinc goes into solution and hydro- gen is evolved. If, on the contrary, the pressure is somewhat increased, hydrogen goes into solu- tion and zinc is precipitiited. That the con- centration of hydrogen gas ( i.e. the amount com- pressed within unit volume) plays the determining role in this phenomenon is clearly evident ; and so this and similar phenomena, too. go to show the inadequacy of older chemical theory, which recognized (in a vague qualitative way) only the 'affinity' factor; and, leaving out of account the factor of concentration, 'explained' the dissolu- tion of metals in acids by the assumption that metals have greater affinity for acid radicles than hydrogen. Electro-Ciiemicai. Series. The potential-dif- ferences established between metals and their salts are now known for a number of dilTerent metals. A detailed explanation of the principles on which they are determined would carry us somewhat beyond the scope of this article. Suf- fice it therefore to state that the .starting-point is a determination, by an extremely ingenious method, of the potentialdilfcrence between metal- lie mercury and solutions of salts, and that the potential-dilTerence between mercury and a nor- mal aqueous solution of one of its salts has thus been shown to be 0.00 volt, the metal being electropositive with regard to the solution. (By a 'normal' solution of a salt is meant a solution containing the molecular weight of the salt in grams. See Souttion). This fact known, there is no difficulty in determining the potent ial- dilTerencc for any other metal ;ind its solutions. Indeed, let the problem be to determine the po- tential difference establislicil when metallic zinc is placed in a normal solution of zinc sulphate. To do this, we may construct a cell consisting of a mercury electrode in a normal solution of
